Quickstart Guide

Get up and running with APIPoints intelligence API in under 5 minutes.

1. Get Your API Key

If you haven't already, create an account and generate an API key from your dashboard.

Your API key looks like: apk_live_...

Important: Store your API key securely. It will not be shown again after generation. Send it with every request as the x-api-key header.

2. Make Your First Request

All intelligence endpoints are accessed via the same base URL and use the x-api-key header for authentication.

Base URL: https://apipoints-worker.francis-e3b.workers.dev

Get LLM Pricing

Retrieve real-time pricing for all tracked models across 8+ providers.

curl -H "x-api-key: YOUR_API_KEY" \
  https://apipoints-worker.francis-e3b.workers.dev/v1/llm-costs

Filter by Provider

curl -H "x-api-key: YOUR_API_KEY" \
  "https://apipoints-worker.francis-e3b.workers.dev/v1/llm-costs?provider=openai"

Get Model Benchmarks

Access MMLU, HumanEval, MATH, GPQA and other benchmark scores.

curl -H "x-api-key: YOUR_API_KEY" \
  https://apipoints-worker.francis-e3b.workers.dev/v1/model-benchmarks

Get Deprecation Notices

See which models are being deprecated and their replacements.

curl -H "x-api-key: YOUR_API_KEY" \
  "https://apipoints-worker.francis-e3b.workers.dev/v1/deprecations?status=announced"

Get Provider Information

curl -H "x-api-key: YOUR_API_KEY" \
  https://apipoints-worker.francis-e3b.workers.dev/v1/providers

Get Recent Changes

Track pricing changes, new models, and deprecation announcements.

curl -H "x-api-key: YOUR_API_KEY" \
  "https://apipoints-worker.francis-e3b.workers.dev/v1/changes?since=2026-06-01"

Get Cost Recommendations

Get model recommendations optimized for your use case.

curl -H "x-api-key: YOUR_API_KEY" \
  "https://apipoints-worker.francis-e3b.workers.dev/v1/recommend?use_case=code"

Calculate Cost

Estimate API costs for a specific model and token volume.

curl -H "x-api-key: YOUR_API_KEY" \
  "https://apipoints-worker.francis-e3b.workers.dev/v1/calculate?model=gpt-4o&input_tokens=500000&output_tokens=50000"

3. API Response Format

All endpoints return JSON with the following structure:

{
  "data": [...],
  "meta": {
    "count": 35,
    "version": "2026.07.24",
    "last_updated": "2026-07-24T00:00:00Z"
  },
  "rate_limit": {
    "remaining": 59,
    "limit": 60
  }
}

4. Available Endpoints

GET /v1/llm-costs — Real-time LLM pricing across all providers

Filters: provider, model, max_input_cost, max_output_cost

GET /v1/model-benchmarks — MMLU, HumanEval, MATH, GPQA scores

Filters: provider, model, min_mmlu, min_human_eval

GET /v1/deprecations — Active and upcoming model deprecations

Filters: provider, status (announced, in_progress, completed)

GET /v1/providers — All tracked LLM providers

Filters: provider (slug)

GET /v1/changes — Recent pricing and model changes

Filters: provider, type, since (YYYY-MM-DD)

GET /v1/recommend — Cost optimization recommendations by use case

Filters: use_case

GET /v1/calculate — Cost calculator for specific model/token combinations

Params: model (required), input_tokens, output_tokens

5. Authentication

APIPoints supports two authentication methods:

API Key (recommended for production): Pass x-api-key: YOUR_API_KEY header
Bearer Token (for dashboard): Pass Authorization: Bearer YOUR_TOKEN header

6. Rate Limits

Rate limits depend on your subscription plan:

Free: 10 requests/minute
Starter ($49/mo): 60 requests/minute
Growth ($149/mo): 300 requests/minute
Enterprise ($499/mo): 1,000 requests/minute

Rate limit headers are included in every response: rate_limit.remaining and rate_limit.limit.

7. MCP Integration

Connect APIPoints to Claude, Cursor, or Windsurf via MCP:

{
  "mcpServers": {
    "apipoints": {
      "command": "npx",
      "args": ["-y", "@apipoints/mcp-server"],
      "env": {
        "APIPOINTS_API_KEY": "YOUR_API_KEY"
      }
    }
  }
}

Once connected, your AI agent can query LLM pricing, benchmarks, and deprecation alerts in real-time.

8. Using in JavaScript/TypeScript

const API_KEY = 'apk_live_...';
const BASE = 'https://apipoints-worker.francis-e3b.workers.dev';

const response = await fetch(`${BASE}/v1/llm-costs?provider=openai`, {
  headers: { 'x-api-key': API_KEY }
});
const data = await response.json();
console.log(data.data); // Array of model pricing objects

9. Using in Python

import requests

API_KEY = "apk_live_..."
BASE = "https://apipoints-worker.francis-e3b.workers.dev"

response = requests.get(
    f"{BASE}/v1/llm-costs",
    headers={"x-api-key": API_KEY}
)
data = response.json()
for model in data["data"]:
    print(f"{model['model']}: ${model['input_cost_per_1m']}/1M input")